Pyridazine-versus pyridine-based tridentate ligands in first-row transition metal complexes

Article abstract of DOI:10.1021/ic200279g

A series of first-row transition metal complexes with the unsymmetrically disubstituted pyridazine ligand picolinaldehyde (6-chloro-3-pyridazinyl) hydrazone (PIPYH), featuring an easily abstractable proton in the backbone, was prepared. Ligand design was inspired by literature-known picolinaldehyde 2-pyridylhydrazone (PAPYH). Reaction of PIPYH with divalent nickel, copper, and zinc nitrates in ethanol led to complexes of the type [Cu^II(PIPYH) (NO₃)₂] (1) or [M(PIPYH)₂](NO₃) ₂ [M = Ni^II (2) or Zn^II (3)]. Complex synthesis in the presence of triethylamine yielded fully- or semideprotonated complexes [Cu^II(PIPY)(NO₃)] (4), [Ni^II(PIPYH)(PIPY)] (NO₃) (5), and [Zn^II(PIPY)₂] (6), respectively. Cobalt(II) nitrate is quantitatively oxidized under the reaction conditions to [Co^III(PIPY)₂](NO₃) (7) in both neutral and basic media. X-ray diffraction analyses reveal a penta- (1) or hexa-coordinated (2, 3, and 7) metal center surrounded by one or two tridentate ligands and, eventually, κ-O,O′ nitrate ions. The solid-state stoichiometry was confirmed by electron impact (EI) and electrospray ionization (ESI) mass spectrometry. The diamagnetic complexes 5 and 6 were subjected to ¹H NMR spectroscopy, suggesting that the ligand to metal ratio remains constant in solution. Electronic properties were analyzed by means of cyclic voltammetry and, in case of copper complexes 1 and 4, also by electron paramagnetic resonance (EPR) spectroscopy, showing increased symmetry upon deprotonation for the latter, which is in accordance with the proposed stoichiometry [Cu ^II(PIPY)(NO₃)]. Protic behavior of the nickel complexes 2 and 5 was investigated by UV/vis spectroscopy, revealing high π-backbonding ability of the PIPYH ligand resulting in an unexpected low acidity of the hydrazone proton in nickel complex 2.

Full text of DOI:10.1021/ic200279g

ARTICLE
pubs.acs.org/IC
Pyridazine- versus Pyridine-Based Tridentate Ligands in First-Row
Transition Metal Complexes
Katrin R. Gr€unwald,† Manuel Volpe,† Pawel Cias,‡ Georg Gescheidt,‡ and Nadia C. M€osch-Zanetti*,†
†Institut f€ur Chemie, Bereich Anorganische Chemie, Karl-Franzens Universit€at Graz, Schubertstrasse 1, A-8010 Graz, Austria
‡Institut f€ur Physikalische und Theoretische Chemie, Technische Universit€at Graz, Stremayrgasse 9/I (A), A-8010 Graz, Austria
S Supporting Information
b
ABSTRACT:
A series of first-row transition metal complexes with the unsymmetrically disubstituted pyridazine ligand picolinaldehyde (6-chloro-
3-pyridazinyl)hydrazone (PIPYH), featuring an easily abstractable proton in the backbone, was prepared. Ligand design was
inspired by literature-known picolinaldehyde 2-pyridylhydrazone (PAPYH). Reaction of PIPYH with divalent nickel, copper, and
zinc nitrates in ethanol led to complexes of the type [CuII(PIPYH)(NO3)2] (1) or [M(PIPYH)2](NO3)2 [M = NiII (2) or ZnII (3)].
Complex synthesis in the presence of triethylamine yielded fully- or semideprotonated complexes [CuII(PIPY)(NO3)] (4),
[NiII(PIPYH)(PIPY)](NO3) (5), and [ZnII(PIPY)2] (6), respectively. Cobalt(II) nitrate is quantitatively oxidized under the
reaction conditions to [CoIII(PIPY)2](NO3) (7) in both neutral and basic media. X-ray diffraction analyses reveal a penta- (1) or
hexa-coordinated (2, 3, and 7) metal center surrounded by one or two tridentate ligands and, eventually, k-O,O0 nitrate ions. The
solid-state stoichiometry was confirmed by electron impact (EI) and electrospray ionization (ESI) mass spectrometry. The
diamagnetic complexes 5 and 6 were subjected to 1H NMR spectroscopy, suggesting that the ligand to metal ratio remains constant
in solution. Electronic properties were analyzed by means of cyclic voltammetry and, in case of copper complexes 1 and 4, also by
electron paramagnetic resonance (EPR) spectroscopy, showing increased symmetry upon deprotonation for the latter, which is in
accordance with the proposed stoichiometry [CuII(PIPY)(NO3)]. Protic behavior of the nickel complexes 2 and 5 was investigated
by UV/vis spectroscopy, revealing high π-backbonding ability of the PIPYH ligand resulting in an unexpected low acidity of the
hydrazone proton in nickel complex 2.
’ INTRODUCTION
binding tendency of the electron-poor second nitrogen donor.
Simple complexes without additional donors or secondary bridg-
ing functionalities featuring pyridazine as a terminal ligand15
display significantly different electronic ligand properties in
comparison to pyridine.
Suitable synthetic pathways are available for symmetric sub-
stitution of 3,6-diformylpyridazine.13 Pyridazines with only one
additional donor next to a ring-nitrogen atom leading to bi- or
tridentate ligands are synthetically more challenging and thus have
been significantly less investigated in coordination chemistry.16ꢀ19
We have recently described an easy entry into pyridazine
chemistry with a methylhydrazone pyridyl substituent next to one
of the nitrogen atoms (picolinaldehyde 3-pyridazinyl-N-methyl-
hydrazone, PIPYMe), with which coordination chemistry of first-
row transition metals was explored.20 Here, the preparation of
an unsubstituted hydrazone ligand [picolinaldehyde (6-chloro-3-
Pyridazines find wide application in drug design1ꢀ3 and as
potential fungicides,4ꢀ6 explaining the development of numer-
ous organic derivatives. In contrast, coordination chemistry of
pyridazines is limited, especially compared to related chemistry
of pyridines. The second nitrogen atom in the heterocyclic system
leads on one hand to a more electron-deficient ring system with
limited σ-donating abilities, which is reflected by the acid
constants (pKa 2.24 for parent pyridazine vs pKa 5.25 for
pyridine), whereas on the other hand effective π-backbonding
is enabled by significantly lowered molecular orbitals. Due to the
different electronic situation in comparison to pyridine systems,
unique chemical behavior can be expected. Pyridazine offers in
principle the possibility to coordinate two metal atoms in
close proximity. However, such behavior is observed only in
pyridazine ligands with preformed compartments or in macro-
cyclic systems.7ꢀ14 In order to accommodate two metals, the
pyridazine ring needs to be substituted in 3- and 6-positions with
two donors to exploit the chelate effect and thus increase the
Received: February 9, 2011
Published: July 15, 2011
r
2011 American Chemical Society
7478
dx.doi.org/10.1021/ic200279g Inorg. Chem. 2011, 50, 7478–7488
|

Inorganic Chemistry
ARTICLE
Scheme 1. Synthesis of Ligand PIPYHa
Figure 1. Ligand system presented in this work in comparison to the
parent PAPYH system.
a (a) Hydrazine (1.6 equiv) in water/NH3, reflux, 3 h.33 (b) 2-Pyridine-
carbaldehyde (1.1 equiv) in EtOH (3 drops of acetic acid), reflux, 2 h.
Here, the syntheses of a series of PIPYH complexes with the
first-row transition metals cobalt, nickel, copper, and zinc are
described. Investigations of their protic behavior, their structural
properties by single-crystal X-ray diffraction analyses, and their
electronic properties by cyclic voltammetry and by electron spin
resonance spectroscopy are compared to the behavior of the
PAPY pyridyl system, revealing interesting differences.
Figure 2. Complexes containing PAPYH with two different metal to
ligand ratios.
pyridazinyl)hydrazone, PIPYH] is described, which is reminis-
cent of the existing picolinaldehyde 2-pyridylhydrazone (PAPYH)
system (Figure 1), allowing direct comparison of a pyridine- and a
pyridazine-based system. Both PIPYH and PAPYH feature a
tautomerizable proton in contrast to the methyl derivative PIPYMe,
which may give rise to interesting properties and applications.
PAPYH, first reported in 1957,21 is a powerful chelating
ligand coordinating in a meridional tridentate mode similar to
terpyridine.22ꢀ26 In contrast, it features a significantly more
flexible hydrazone bridge and a secondary amine functionality
with a proton in the ligand backbone whose pronounced acidity
increases upon coordination.22,27,28 Thus, it can be easily re-
moved with a base and readded with an acid, reversibly. The
deprotonated complexes are readily soluble in a variety of organic
solvents and were found to be intensely colored, rendering them
suitable acidꢀbase indicators.28,29 Due to their extraordinary
stability toward acids and bases, apart from de- or reprotonation,
PAPYH/PAPYꢀ solutions allow for selective extraction of
various metals from aqueous media.30,31
’ RESULTS AND DISCUSSION
Synthesis of Ligand. Ligand PIPYH was synthesized in a two-
step procedure as shown in Scheme 1. Substitution of one
chlorine of 3,6-dichloropyridazine with hydrazine proceeded
smoothly in boiling aqueous ammonia, yielding a white product
1
in high purity. Monosubstitution was evidenced by H NMR
spectroscopy, as the singlet assignable to the equivalent aromatic
protons of the starting material at 7.51 ppm splits into two
doublets at 7.26 and 7.13 ppm. Two broad signals at 6.62 and
2.78 ppm in the ratio of 1:2 for one secondary NH and two
primary NH2 protons appear in the spectrum. Subsequent Schiff-
base condensation employing 2-pyridinecarbaldehyde in ethanol
gave the ligand in the form of white needles. Imine formation is
evidenced by a new singlet at 8.16 ppm in the 1H NMR spectrum.
Two shifted pyridazine doublets at 7.76 and 7.71 ppm and the
shifted broad resonance at 11.97 ppm for the acidic NH proton
indicate quantitative formation of PIPYH.
Synthesis of Complexes. The coordinating ability of PIPYH
toward the first-row transition metals nickel, copper, and zinc was
investigated. Addition of an ethanol solution of the respective
metal salt [Ni(NO3)2 6H2O, Cu(NO3)2 3H2O, or Zn-
Coordination chemistry of PAPYH has been thoroughly
investigated. The tridentate ligand coordinates strongly to diva-
lent metal ions, forming octahedral complexes of the type [M
3
3
(PAPYH)2]X2 (e.g., X = ClOh or PFh) or square pyramidal
4
6
(NO3)2 4H2O] to a boiling ethanol solution of PIPYH led in
3
complexes containing only one PAPYH molecule [M
(PAPYH)Y2] (e.g., Y = Br or Cl) depending on the counterion
(Figure 2).24,27
all cases to an immediate color change from pale yellow to brown.
After cooling to room temperature and slow evaporation of the
solvent in all three reaction solutions, crystals suitable for X-ray
diffraction analysis were obtained. For copper, determination of
the molecular structure revealed the formation of complex
[CuII(PIPYH)(NO3)2] (1) with 1 equiv of the hydrazone ligand
and two coordinated nitrate ligands. The direct inlet electron
impact mass spectrometry (EI-MS) spectrum of solid 1 shows
[Cu(PIPY)(NO3)]+. This is consistent with the literature, where
copper complexes of 1:1 stoichiometry are also reported for
PAPYH, especially when coordinating counterions are
present.24,25 However, titration studies of the PAPY complexes
showed that, in solution, various complexes with different
metal to ligand ratios are present.32 In agreement with that, the
Among late first-row transition metals, cobalt(II) salts behave
differently. Due to the relative instability of the divalent state, CoII
is readily oxidized to CoIII by PAPYH, presumably under forma-
tion of molecular hydrogen.32 The resulting semiprotonated 2:1
complex [CoIII(PAPYH)(PAPY)]2+ is readily fully deprotonated,
for example, by recrystallization from aqueous ethanol.23
The deprotonated PAPYꢀ anion is fully aromatic, which leads
to a distribution of the formal charge at the hydrazone nitrogen
all over the molecule. Detailed studies on the location of the
charge in the PAPYH system or in related frameworks have not
been reported yet.
7479
dx.doi.org/10.1021/ic200279g |Inorg. Chem. 2011, 50, 7478–7488

Inorganic Chemistry
ARTICLE
Scheme 2. Formation of Complexes 1ꢀ6 Employing Neutral PIPYH or Monoanionic PIPYꢀ
electrospray ionization mass spectometry (ESI-MS) spectrum of
1 in acetonitrile seems to show the presence of several species.
In the case of nickel and zinc, dicationic complexes of the type
[M(PIPYH)2](NO3)2 [M = NiII (2) or ZnII (3)] were formed
regardless of the employed stoichiometry (Scheme 2). This was
further confirmed by direct inlet EI-MS analyses of solid samples
of 2 and 3. Solution ESI mass spectrometry of complex 2 shows
the [M(PIPYH)2 ꢀ H]+ peak as well. All complexes feature
tridentate ligand molecules, which coordinate in a meridional
fashion. Although the complexes possess low solubility in
common organic solvents, the concentration was enough for
recording CV data. In the case of zinc complex 3, dissolution in
CD3CN caused decomposition of the complex, evidenced by the
appearance of free ligand signals in the 1H NMR spectrum.
Deprotonated Complexes. Addition of PIPYH to the metal
nitrate solution in the presence of 1 equiv of triethylamine led, in
the case of copper and nickel nitrate, to intensely colored
crystalline precipitates (Scheme 2). With copper, we believe that
the deprotonated complex [CuII(PIPY)(NO3)] (4) was formed,
as evidenced by elemental analysis. Direct inlet EI-MS of the solid
sample shows the molecular ion [Cu(PIPY)]+ with the correct
isotopic pattern, whereas a peak for [Cu(PIPY)2]+ is consistently
absent, identical to the data found for 1. According to ESI-MS in
acetonitrile, again, several species seem to be present in solution,
in agreement with the behavior seen for related PAPY
complexes.32 In both cases, mass spectrometry is unsuitable to
distinguish between compounds of types [Cu(LH)(NO3)2] and
[Cu(L)(NO3)]. For this reason, compounds 1 and 4 were
further investigated by electron paramagnetic resonance (EPR)
spectroscopy supporting their formulation (vide infra).
indicates identical L:M = 2:1 ratios for both complexes. Further-
more, the absence of a molecular ion for [Ni(PIPY)2]+, together
with the elemental analysis data, points to the formation of a
semideprotonated compound rather than [Ni(PIPY)2].
The reaction of zinc nitrate with a mixture of PIPYH and NEt3
in ethanol yielded a yellow clear solution after cooling to room
temperature. Evaporation of the solvent gave a yellow suspen-
sion, from which a waxy solid was isolated by centrifugation and
washed with diethyl ether. The material is well soluble in ethanol
but only slightly in acetonitrile. For reasons of comparison, a 1H
NMR spectrum was recorded in CD3CN, although an overnight
measurement was necessary (Figure 3). It shows one set of
resonances assignable to the protons of the deprotonated ligand
due to the absence of an NꢀH resonance. All aromatic signals are
significantly shifted compared to those of PIPYH. For this
reason, we believe that the symmetric doubly deprotonated
complex [Zn(PIPY)2] (6) is formed. Furthermore, EI-MS shows
a peak with the expected isotopic pattern for [Zn(PIPY)2]+. The
upfield shift of the aromatic resonances is presumably due to a
certain degree of metal-to-ligand π-backbonding.
Reactivity toward a Cobalt(II) Salt. The reaction of Co-
(NO3)2 6H2O with 1 equiv of PIPYH in the presence as well as
3
in absence of triethylamine led to isolation of the same product
(Scheme 3), in contrast to the above-described behavior. The
ligand solution in boiling ethanol with or without triethylamine
turned immediately deep purple upon addition of the metal salt
solution. After cooling to room temperature, a brown solid was
formed upon removal of the solvent. When this material was
recrystallized from ethanol, small dark green single crystals
started to grow as the solvent slowly evaporated overnight.
Redissolution of these crystals in polar solvents such as dimethyl
sulfoxide (DMSO), acetonitrile, methanol, or ethanol reinstated
the intensively purple-colored solution. The obtained crystals
proved to be suitable for X-ray diffraction analysis, which
revealed the formation of complex [CoIII(PIPY)2](NO3) (7)
with two coordinated hydrazone ligands and 1.5 molecules of
ethanol in the unit cell. The presence of the latter probably
With nickel, the semideprotonated compound [NiII(PIPYH)
(PIPY)](NO3) (5) is presumably formed. Also, the two nickel
complexes 2 and 5 show identical solid-state EI-MS data, the
molecular ion being [Ni(PIPYH)(PIPY)]+. In contrast to the
copper complexes 1 and 4, complex 5 seems more stable toward
ligand substitution as the ESI mass spectrum in acetronitrile
also shows the highest mass to be [Ni(PIPYH)(PIPY)]+. This
7480
dx.doi.org/10.1021/ic200279g |Inorg. Chem. 2011, 50, 7478–7488

Inorganic Chemistry
ARTICLE
Figure 3. 1H NMR spectra (aromatic region in CD3CN) of ligand PIPYH and complex [Zn(PIPY)2] (6): (a) pyridazine, (b) imine, and (c) pyridyl
proton resonances.
Scheme 3. Formation of Cobalt Complex 7 Employing PIPYH in the Presence and Absence of NEt3
justifies the different color of the crystals compared to the
isolated compound, a brown amorphous powder. One nitrate
counterion was found in the unit cell, pointing to the oxidation of
CoII to CoIII. This fact is further supported by the CoꢀN bond
distances, which fall in the range reported for CoIIIꢀN, whereas
significantly longer distances would be expected in the case of the
1
divalent state.34 In the H NMR spectrum, complex 7 exhibits
sharp resonances in CD3CN as expected for the CoIII d6 low-spin
1
center. H NMR spectra of the purple acetonitrile solutions of
the brown material and the green crystals are identical and show
one set of signals for a coordinated PIPYꢀ ligand. In the spectrum
of the green crystals, ethanol was found, which integrates for 1.5
molecules per cobalt. Due to poor solubility of the complex, no
Figure 4. Molecular view of complex [CuII(PIPYH)(NO3)2] (1)
including a hydrogen-bonded nitrate ion of the neighboring complex
13C NMR spectra could be obtained even after prolonged
measurement. Elemental analysis of both materials confirms
the composition and points to identical complexes. Furthermore,
EI mass spectrometry shows the formation of the [Co(PIPY)2]+
core for both materials. Deprotonated PAPYꢀ complexes show
high solubilities in organic solvents including benzene and
chloroform,27 which is why they have been proposed for ion
extraction applications from water.30,31 In contrast to that, the
solubility of cobalt complex 7 is dramatically decreased, with it
being soluble only in polar solvents such as ethanol or acetonitrile
and completely insoluble in tetrahydrofuran (THF) or ethers.
Similar to the behavior of PAPYH, the cobalt ion in 7 is
presumably oxidized by the PIPYH ligand, consistent with
oxidizing ability of PAPYH solutions toward CoII ions, which
was reported early on.21,32 When the reaction was performed
with rigorously degassed solvent under an Ar atmosphere, the
identical diamagnetic compound was isolated in comparable
ion (50% probability). Hydrogen atoms (except the hydrazone proton in
the ligand backbone) have been omitted for clarity.
yields (64% based on the ligand). Layering an ethanol solution
with dry and degassed diethyl ether in an argon atmosphere
afforded green single crystals that proved to exhibit identical
crystallographic parameters to those of crystals obtained under
ambient conditions. Thus, oxygen can be excluded as the oxidizing
agent and the ligand seems more likely to provide the redox
equivalents by forming hydrogen. Furthermore, the reduction
potential of CoIII f CoII (ꢀ0.866 V vs Ag/Ag+ or ꢀ0.621 V vs
normal hydrogen electrode, NHE) points toward a spontaneous
oxidation of the metal in the presence of protons.
Molecular Structures of Complexes. For all protonated com-
plexes 1ꢀ3 as well as for complex 7, single crystals suitable for
crystallographic analyses could be obtained by slow evaporation of
7481
dx.doi.org/10.1021/ic200279g |Inorg. Chem. 2011, 50, 7478–7488

Inorganic Chemistry
ARTICLE
Table 2. Selected Angles for Complexes 1ꢀ3 and 7
Table 1. Selected Bond Lengths for Complexes 1ꢀ3 and 7
bond angle, deg
bond length, Å
2a
3
7b
angle
1
2a
3
7b
bond
1
N2ꢀMꢀN4
N2ꢀMꢀN5
N2ꢀMꢀN7
N2ꢀMꢀN9
N2ꢀMꢀN10
N4ꢀMꢀN5
N4ꢀMꢀN7
N4ꢀMꢀN9
N4ꢀMꢀN10
N2ꢀMꢀO1
N2ꢀMꢀO4
N4ꢀMꢀO1
N4ꢀMꢀO4
N5ꢀMꢀO1
N5ꢀMꢀO4
O1ꢀMꢀO4
N5ꢀMꢀN7
N5ꢀMꢀN9
N5ꢀMꢀN10
N7ꢀMꢀN9
N7ꢀMꢀN10
N9ꢀMꢀN10
78.99(4)
76.60(6)
154.30(6)
96.58(9)
73.44(4)
147.44(4)
94.30(4)
121.47(4)
90.74(4)
74.33(4)
102.78(4)
164.22(4)
112.82(4)
81.0(3)
164.3(3)
91.1(3)
94.7(3)
91.1(3)
83.3(3)
99.9(3)
175.8(3)
96.3(3)
MꢀN2
MꢀN4
MꢀN5
MꢀO1
MꢀO4
MꢀN7
MꢀN9
MꢀN10
2.0017(9)
1.9578(10)
1.9958(9)
1.9458(8)
2.2731(9)
2.1037(17)
2.0027(17)
2.1177(17)
2.1803(11)
2.0992(11)
2.2247(11)
1.901(7)
1.875(8)
1.948(7)
157.61(4)
102.80(7)
91.26(7)
80.14(4)
77.81(7)
2.1037(17)
2.0027(17)
2.1177(17)
2.1493(11)
2.1533(11)
2.1333(11)
1.904(7)
1.876(8)
1.937(8)
102.80(7)
179.12(10)
102.82(7)
a Second ligand molecule generated by symmetry operation {#1 (ꢀx +
2, y, ꢀz + /2); #2 (ꢀx + 2, y, ꢀz + /2}: N2i, N4i, and N5i are
represented by N7, N9, and N10. b Data for only one molecule of the two
found in the asymmetric unit are given.
1
3
100.81(4)
85.05(4)
174.59(4)
100.10(4)
99.19(4)
106.60(4)
85.24(4)
the solvent. Molecular views of complexes 1, 2 and 7 are displayed
in Figures 4ꢀ6. A representation of complex 3 is included in the
Supporting Information (Figure S1). Selected bond lengths and
angles are given in Tables 1 and 2, and crystallographic data are
listed in Table 3.
91.26(7)
102.82(7)
92.17(9)
76.60(6)
154.30(6)
77.81(7)
97.13(4)
91.06(4)
97.51(4)
72.63(4)
143.98(4)
74.37(4)
90.2(3)
100.9(3)
92.0(3)
80.4(3)
163.8(3)
83.4(3)
In the copper complex [CuII(PIPYH)(NO3)2] (1) (Figure 4),
the metal atom is coordinated by one neutral ligand and two
nitrate ions to form a distorted square pyramidal geometry. The
PIPYH ligand molecule is coordinated in a tridentate meridional
mode to occupy three positions of the basal plane. The fourth
corner and the apical position are taken by the oxygen atoms of
two nitrates, which coordinate as terminal k1-O ligands. Similar
square pyramidal configurations have previously been observed
for copper halide complexes bearing also one PAPYH ligand.24,25
O2 is approaching the second apical position of an octahedron,
but the long atom distance [2.734(9) Å] implies only a weak
interaction. The Cu1ꢀN4 bond is the shortest metalꢀligand
distance [1.9578(10) Å] and is significantly shorter than Cu1ꢀN2
and Cu1ꢀN5, which are 2.0017(9) and 1.9958(9) Å, respec-
tively, which is in perfect agreement with literature data of
PAPYH copper(II) complexes.24,25 The apical nitrate oxygen
distance to the metal center is longer [2.2731(9) Å] than the
basal nitrate oxygen [1.9548(8) Å], which represents the shortest
donorꢀacceptor distance of all. Distorted square pyramidal
geometry is reflected by the angles N2ꢀCu1ꢀN5 [157.61
(4)ꢀ], N4ꢀCu1ꢀO1 [174.59(4)ꢀ], and N4ꢀCu1ꢀO4 [100.10
(4)ꢀ], which deviate from ideal geometry. The acidic hydrogen
atom could be located from the difference map as it forms a
hydrogen bond to a nitrate oxygen atom of the neighboring
a Second ligand molecule generated by symmetry operation {#1 (ꢀx +
1
3
2, y, ꢀz + /2); #2 (ꢀx + 2, y, ꢀz + /2}: N2i, N4i, and N5i are
represented by N7, N9, and N10. b Data for only one molecule of the two
found in the asymmetric unit are given.
94.30(4)ꢀ for 3]. All acidic hydrogen atoms could be located and
were found to form hydrogen bonds to the nitrate counterions.
For the cobalt complex [CoIII(PIPY)2](NO3) (7), two com-
plex molecules were found in the asymmetric unit, showing only
small deviations in their octahedral geometries. A representative
molecule is shown in Figure 6. The metal atom is coordinated by
two ligand molecules in a mer-tridentate fashion, forming an
octahedral cation. The nitrate counterion is not coordinated.
Metalꢀnitrogen bond lengths range from 1.875(8) to 1.948(7) Å
and are significantly shorter than for complexes 1ꢀ3 due to the
decreased ionic radius as well as the increased electrophilicity of
CoIII with respect to CoII. All bond distances match well with
reported values for CoIIIꢀN and are significantly shorter than
expected values for the case of the divalent state.34 The bond
lengths were found to be shortest for the metalꢀimino nitrogen
atoms N4 and N9 and the longest for the metalꢀpyridine
nitrogen atoms N5 and N10. The same trend was also found
for compounds 2 and 3.
AcidꢀBase Behavior. The reactivity of nickel complex
[NiII(PIPYH)2](NO3)2 (2) toward bases and of [NiII(PIPYH)
(PIPY)](NO3) (5) toward acids was investigated by UV/vis
spectroscopy. To a sample of the respective compound in dry
acetonitrile was added up to 100 equiv of the base or acid. After
the addition of 5 equiv of base or acid, respectively, a UV/vis
spectrum was recorded. Surprisingly, compound 2 seems to be
unreactive toward deprotonation, as the spectrum did not show
significant changes after addition of either triethylamine or
sodium hydroxide. Only a slight hypsochromic shift is apparent,
complex molecule [H3 O5 1.879(9) Å].
3 3 3
The two complexes [NiII(PIPYH)2](NO3)2 (2) (Figure 5)
and [ZnII(PIPYH)2](NO3)2 (3) are isostructural. The metal
atom is coordinated by two tridentate PIPYH ligands, leading to
a distorted octahedral geometry. In general, all metalꢀligand
bonds are slightly shorter for the nickel complex 2 than for the
zinc complex 3 due to the smaller ionic radius. All metalꢀligand
bond distances are in a similar range from 2.0027(17) to
2.1177(17) Å for 2 and 2.0992(11) to 2.1533(11) Å for 3, which
is in agreement with literature reported data.35ꢀ37 Distorted
octahedral geometry is evidenced by deviation of all the involved
angles from ideal geometries [e.g., N2ꢀMꢀN5 is 154.30(6)ꢀ for
2 and 147.44(4)ꢀ for 3, or N2ꢀMꢀN7 is 96.58(9)ꢀ for 2 and
7482
dx.doi.org/10.1021/ic200279g |Inorg. Chem. 2011, 50, 7478–7488

Inorganic Chemistry
ARTICLE
Table 3. Crystallographic Data and Structure Refinement for Complexes 1-3 and 7
1
2 EtOH
3 EtOH
7 1.5EtOH
3
3
3
empirical formula
Mr, g/mol
CuC10H8ClN7O6
421.22
NiC24H32Cl2N12O7
730.23
ZnC22H22Cl2N12O7
702.79
CoC23H23Cl2N11O4.5
655.35
crystal description
crystal system
space group
a, Å
green column
monoclinic
P2(1)/n
8.2730(3)
14.7159(5)
12.5797(4)
90.00
orange column
monoclinic
C2/c
yellow cube
triclinic
black block
triclinic
P1
P1
19.3589(17)
10.2495(9)
14.6535(14)
90.00
9.1327(5)
10.8382(5)
14.6431(7)
98.457(2)
90.961(3)
97.988(2)
1418.71(12)
2
8.9527(8)
15.3964(13)
20.5716(18)
80.312(5)
84.071(5)
83.546(5)
2767.1(4)
4
b, Å
c, Å
R, deg
β, deg
104.5530(10)
90.00
103.300(4)
90.00
γ, deg
volume, Å3
1482.37(9)
4
2829.5(4)
4
Z
T, K
100(2)
100(2)
100(2)
100(2)
Dc, g/cm3
1.887
1.714
1.645
1.573
μ(MoKR), mmꢀ1
1.703
0.945
1.120
0.868
F(000)
844
1512
716
1340
reflections collected
unique reflns
reflns with I g 2σ(I)
R(int), R(σ)
no. of param/restraints
final R1 a, wR2 b (I g 2σ)
R indices (all data)
GOF on F2
116698
36659
127860
26117
4752
4784
8988
12155
4460
4159
8327
6186
0.0355, 0.0116
226/0
0.0351, 0.0245
186/0
0.0326, 0.0144
399/0
0.0677, 0.1514
703/31
0.0214, 0.0580
0.0237, 0.0593
1.099
0.0327, 0.0820
0.0399, 0.0854
1.074
0.0294, 0.0761
0.0330, 0.0788
1.070
0.0890, 0.2207
0.1328, 0.2496
0.950
largest diff. peak and hole, e/Å3
0.554, ꢀ0.393
0.691, ꢀ0.629
810788
1.035, ꢀ0.613
810790
1.512, ꢀ1.594
810791
CCDC deposition no.
a R1 = Σ Fo| ꢀ |Fc /Σ|Fo|. b wR2 = {Σ[w(Fo ꢀ Fc2)2]2/Σ[w(Fo ) }
810789
2
2 2 1/2
.
Figure 6. Molecular view of complex [CoIII(PIPY)2](NO3) (7) (50%
probability). Hydrogen atoms have been omitted for clarity. Only one
molecule of the asymmetric unit is displayed.
protonated by weak acids. Figure 7 displays UV/vis spectra of the
reaction of 5 with acetic acid (100 equiv). Comparison with the
spectrum of isolated 2 (inset, Figure 7) shows the course of the
protonation process; however, a huge excess is required to
achieve full conversion. Unfortunately, stronger acids led to
decomposition of the compound, as indicated by the appearance
of absorptions in the UV/vis spectrum characteristic for the
uncoordinated PIPYH ligand and the disappearance of the
absorptions shown in Figure 7. Analogous PAPY complexes,
however, are extraordinarily stable toward strong acids or bases
Figure 5. Molecular view of complex [NiII(PIPYH)2](NO3)2 (2) (50%
probability). Hydrogen atoms (except the hydrazone proton in the
ligand backbone) and solvent molecules have been omitted for clarity.
Hydrogen bonding is indicated by a dashed bond.
due to altered solvent properties, but the intensities of the peaks
remain unchanged. It is not clear why compound 5 can easily be
obtained in situ but not by a stepwise procedure from isolated 2
and subsequent addition of base. However, compound 5 can be
7483
dx.doi.org/10.1021/ic200279g |Inorg. Chem. 2011, 50, 7478–7488

Inorganic Chemistry
ARTICLE
and can be de- and reprotonated reversibly, rendering them
suitable titration indicator substances.28,29
Similar protonation and proton abstraction experiments
could not be performed with the two copper complexes [CuII
(PIPYH)(NO3)2] (1) and [CuII(PIPY)(NO3)] (4) due to the
fact that in solution several species seem to be present, as
observed by ESI-MS.
assigned to the redox couple NiII f NiI (Figure 8a). The
negative shift of the reduction potential is caused by the ligand
electron-donation ability and is similar to the potential of related
nickel complexes featuring also N6 coordination environ-
ments.39,40 A higher electron density due to one anionic ligand
is anticipated at the metal core of complex 5, leading to a more
negative reduction potential. However, the opposite trend is
observed; thus the increased σ-donating ability of the deproto-
nated ligand is supposed to be compensated for by a significant
π-acceptor contribution.
As expected, similar experiments with the CoIII complex 7
showed it to be unreactive toward protonation. Weak acids such
as acetic acid did not affect the UV/vis spectrum, whereas strong
acids like nitric acid, p-toluenesulfonic acid, or hydrochloric acid
led to full decomposition of the compound. These observations
propose a decreased coordination capability of the ligand upon
introduction of an additional nitrogen atom in the heterocycle so
that protonation of the backbone starts competing with displace-
ment of the ligand.
Cobalt complex 7 (Figure 8b) exhibits two reversible redox
waves, which are assigned to the consecutive CoIII f CoII (E1/2
ꢀ0.866 V) and CoII f CoI (E1/2 ꢀ1.602 V) redox processes,
giving ample evidence for 7 being a CoIII complex.
EPR Spectroscopy and Calculations of Complexes 1 and 4.
The EPR spectra obtained from powders of [CuII(PIPYH)
(NO3)2] (1) and [CuII(PIPY)(NO3)] (4) were well distinguish-
able but barely resolved. Since an unambiguous parameter set for
the simulation of the experimental spectra was hardly attainable,
we decided to perform quantum-mechanical calculations at
appropriate levels of theory to determine the geometries of the
complexes and, on the basis of these data, we evaluated the EPR
parameters (g and hyperfine tensors). From the calculated EPR
data, the EPR spectra were calculated and compared to the
experimental counterparts.
The basis of our calculations was the X-ray geometry of
[CuII(PIPYH)(NO3)2] (1), which we have chosen as the start-
ing point of the computations. The initial, experimentally
determined geometry was optimized by density functional theory
(DFT) at the B3LYP/TZVP level of theory. The calculated
geometry essentially resembles the arrangement of the ligands
Electrochemical Studies. Redox properties of complexes 1, 2,
4, 5, and 7 were investigated by cyclic voltammetry in dry
acetonitrile solution, and the data are summarized in Table 4.
Copper complexes 1 and 4 show one pseudoreversible redox
couple each, with cathodic reductions at ꢀ0.920 and ꢀ0.875 V,
respectively, as well as oxidative responses with narrow widths
and high peak currents at ꢀ0.615 and ꢀ0.644 V, which are
typical for anodic stripping of copper metal.38 Thus, no further
insight into potentially present species in solution could be
obtained by CV experiments.
Nickel complexes 2 and 5 show two irreversible reduction
waves with Epc at ꢀ1.344 and ꢀ1.225 V, respectively, which were
Table 4. Electrochemical Data for 1, 2, 4, 5, and 7 in Dry
Acetonitrile Solution at Room Temperaturea
compd
Epc, V
Epa, V
ΔEp, mV
E1/2, V
redox process
1
4
2
5
7
ꢀ0.920
ꢀ0.875
ꢀ1.365
ꢀ1.203
ꢀ0.902
ꢀ1.643
ꢀ0.615
ꢀ0.644
305
231
ꢀ0.768
ꢀ0.760
CuII f Cu
CuII f Cu
NiII f NiI
NiII f NiI
CoIII f CoII
CoII f CoI
ꢀ0.830
ꢀ1.561
72
82
ꢀ0.866
ꢀ1.602
a Measured by CV in CH3CN at 50 mV/s. E vs Ag/AgNO3. Conditions:
Pt disk (working) and [Ag+/AgNO3 0.01 M in CH3CN, 0.1 M
[nBu4N][PF6]] (reference) electrodes, supporting electrolyte 0.05 M
TBAP solution in CH3CN in an Ar atmosphere.
Figure 7. UV/vis spectra of the titration of 5 with 100 equiv of acetic
acid in dry acetonitrile (5 equiv per spectrum). (Inset) Spectrum of
isolated 2 in dry acetonitrile.
Figure 8. Cyclic voltammogram of complexes 2, 5, and 7 in acetonitrile solution. Scan rate 50 mV/s.
7484
dx.doi.org/10.1021/ic200279g |Inorg. Chem. 2011, 50, 7478–7488

Inorganic Chemistry
ARTICLE
Table 5. Calculated and Experimental g and A Tensors of 1
and 4
experimentally and computationally determined geometry dis-
play common features, leading to a reasonable fit with the
experimental EPR spectrum. Using these data for the simulation
of the EPR spectrum indicates a reasonable agreement with its
experimental counterparts.
gxx
gyy
gzz
Axx
Ayy
Azz
1 (calc/calc geom)a
1 (calc/exp geom.)a 2.044
2.008
2.159 2.200 32.8 164.0
405.3
536.4
ꢀ
The NH proton of the PIPYH ligand and one NO3 anion
2.066 2.156 13.5
2.112 2.164
92.2
22.2
were removed from 1 and subsequently this complex, [CuII
(PIPY)(NO3)] (4) was optimized as described above.
1 (exp)
4 (calc geom)b
2.069
2.042
1.992
2.055 2.151 18.8
538.3
The calculated g and A tensor components of 4 reflect an
essentially axial symmetry. With g^ = 2.048 [(gxx + gyy)/2] and
g = 2.151 together with the hyperfine parameters A^ and A =
20.5 and 538 MHz, respectively, a nicely matching simulation of
the experimental EPR spectrum can be accomplished (Figure 9).
Electron Distributions in 1 and 4. The isosurface plots of the
spin density distributions in 1 and 4 were computed at the same
level of theory. Calculations show that, in the case of 1, the spin
population is localized mainly on the metal center and nitrate
ions. Only a very low portion of the spin is present on the
nitrogen atoms coordinated directly to the CuII center. Presum-
ably the “real” spin distribution leaves a higher portion of the
negative charge within the π system of the PIPYH ligand, owing
4 (exp)
1.992 2.318
a The EPR parameters were calculated on the basis of the computation-
ally optimized geometry (first line) and the X-ray data (second line) of 1
b No X-ray structure available.
ꢀ
to the attenuation of the charge density at the NO3 groups,
which is caused by intermolecular hydrogen bonding (see above
and Figure 4). On the other hand, in the deprotonated complex
4 the spin is distributed more uniformly between CuII, the
NOh counterion, and the nitrogen atoms of the PIPY ligand
3
(Figure 10).
Figure 9. EPR spectra of 1 and 4 and their simulated counterparts. The
simulations are accomplished with the use of calculated data presented
in Table 5.
’ CONCLUSIONS
Two series of protonated and deprotonated complexes of NiII,
CuII, and ZnII have been synthesized and characterized by NMR,
attenuated total reflection infrared (ATR-IR) and UV/vis and
mass spectroscopy as well as cyclic voltammetry. Two stoichio-
metries were observed for the protonated complexes by X-ray
crystallography, namely, ligand:metal = 1:1 for copper (1) and
2:1 for nickel (2) and zinc (3). The identical stoichiometries for
the deprotonated counterparts in the case of copper (4), nickel
(5), and zinc (6) were strongly indicated by EI-MS measure-
ments. The same metal-induced preference for equimolar co-
ordination was previously described for the PAPYH/PAPYꢀ
system.21,32
The semideprotonated nickel complex 5 could irreversibly be
fully protonated by weak acids. ESI-MS evidenced intact com-
plexes in solution; thus the enhanced proton affinity is tentatively
ascribed to an increased π-backbonding of the ligand, which
increases significantly the electron density on the ligand. These
observations were supported by reduction potentials of com-
plexes 2 and 5 obtained by CV. The ligand itself is readily
deprotonated by triethylamine during complex synthesis. With
the PAPHY/PAPYꢀ system, no π-backbonding is reported.
Coordination of the latter leads solely to σ-donation to the
metal, which significantly increases the acidity of the abstractable
hydrazone proton. The copper complexes 1 and 4 appeared to be
unstable in solution as evidenced by ESI-MS; thus comparative
studies could not be carried out.
For cobalt, only one type of fully deprotonated complex could
be obtained, which shows a diamagnetic CoIII core. In the
literature, the same behavior was described for the PAPYH/
PAPYꢀ system.21,32 CoII is presumably oxidized by the proton of
the acidic ligand, which is reduced to H2. This is supported by the
fact that the same product is obtained in the absence of oxygen.
Figure 10. Spin density distributions in 1 and 4 computed at B3LYP/
TZVP level (( 0.0008 au isosurfaces; positive spin density is indicated in
red, negative in green).
established by X-ray structure analysis. Remarkably, however, the
distances between the CuII center and the adjacent ligand N
atoms are markedly longer in the calculation (see Supporting
Information) than those determined by experiment. This can be
explained by the experimentally observed hydrogen bonding
between the (acidic) NH group of the PIPYH ligand and
ꢀ
adjacent NO3 counterions (Figure 4). The EPR spectrum of
1 possesses no marked parallel features and a rather broad signal
in the perpendicular domain. The signal could point either to
distorted axial symmetry with a substantially broadened parallel
region or to a substantially lower symmetry. Some more insight is
provided by the calculation. It suggests a clearly nonaxial g and
A tensor. The EPR parameters were calculated, on the basis of
both computationally optimized geometry and, for 1, also X-ray
data. The computed EPR parameters (Table 5) based on the
7485
dx.doi.org/10.1021/ic200279g |Inorg. Chem. 2011, 50, 7478–7488

Inorganic Chemistry
ARTICLE
During complex synthesis, the acidity of the proton in the
backbone is further increased by the enhanced electron density
on the oxidized metal and it is lost easily. Thus, no reprotonation
is feasible, as with weak acids no reaction is observed, but strong
acids lead to immediate decomposition of the compound.
In the case of CuII complex 1 and its deprotonated derivative
4, EPR spectroscopy and DFT calculations reveal a rather
increased symmetry upon deprotonation. This becomes evident
from the transition of the nonaxial g tensor for 1 to an essentially
axial one for 4 and is in line with the presence of only one tightly
bound NO3ꢀ counterion. With the decreased number of ligands,
the overall symmetry is enlarged. This observation corresponds
to the elimination of ꢀhydrogen bonding between the ligand
and neighboring NO3 counterions that is present in [CuII
(PIPYH)(NO3)2] (1) (Figure 4).
General Procedure A (Synthesis of Complexes 1ꢀ3). To a
boiling solution of PIPYH in ethanol was added 1 equiv of the metal salt
[Ni(NO3)2 6H2O, Cu(NO3)2 3H2O, or Zn(NO3)2 4H2O] as a 0.1 M
solution in ethanol. A color change was immediately apparent. The
reaction mixtures were heated to reflux for an additional 20 min,
followed by cooling to room temperature. The products crystallized
overnight at ambient temperature by slow evaporation of the solvent.
The obtained crystals were isolated by filtration and dried in vacuo.
Synthesis of [CuII(PIPYH)(NO3)2] (1). General procedure A was fol-
3
3
3
lowed by employing PIPYH (50 mg, 0.21 mmol) and Cu(NO3)2 3H2O
3
(0.21 mmol, 2.1 mL of a 0.1 M solution in ethanol) in 10 mL of ethanol;
compound 1 was obtained as green parallelepipeds (45 mg, 51%) by
crystallization at room temperature overnight. IR (ATR, cmꢀ1) ν = 1594
(w), 1486 (m), 1283 (s), 1010 (m), 930 (m), 778 (s), 417 (s). EI-MS (m/
z) 295 [M ꢀ 2NO3 ꢀ H]+. Anal. Calcd for CuC10H8N7O6Cl(Found): C,
28.52 (28.77); H, 1.91 (1.59); N, 23.28 (23.15).
Synthesis of [NiII(PIPYH)2](NO3)2 (2). General procedure A was
followed by employing PIPYH (50 mg, 0.21 mmol) and Ni-
’ EXPERIMENTAL SECTION
(NO3)2 6H2O (0.21 mmol, 2.1 mL of a 0.1 M solution in ethanol) in
3
10 mL of ethanol; compound 2 was obtained as orange rectangular rods
(49 mg, 72% based on ligand) by crystallization at room temperature
overnight. IR (ATR, cmꢀ1) ν = 1594 (m), 1420 (s), 1136 (s), 897 (s),
839 (s), 770 (s), 640 (s), 571 (s), 418 (m). MS (ESI in CH3CN, m/z)
General. All chemicals were purchased from commercial sources
and used without further purification.
Spectroscopy. All NMR spectra were measured on a Bruker
Avance III spectrometer (300 MHz for 1H and 75 MHz for 13C NMR).
The 1H NMR spectroscopic data are reported as s = singlet, d = doublet,
t = triplet, m = multiplet or unresolved, br = broad signal, coupling
constant(s) in hertz, shifts in parts per million (ppm) relative to the
solvent residual peak. UV titrations were performed on a Varian Cary 50
spectrophotometer with a thermostated cuvette cavity. Infrared spectra
were measured on a Bruker ALPHA-P Diamant ATR-FTIR spectro-
meter. Mass analysis was performed on an Agilent Technologies 5975C
inert XL MSD direct insertion probe spectrometer applying electron
impact ionization.
525 [M ꢀ H]+. Anal. Calcd for NiC20H16N12O6Cl2 H2O (Found): C,
3
35.96 (35.89); H, 2.72 (2.56); N, 25.16 (25.41).
Synthesis of [ZnII(PIPYH)2](NO3)2 (3). General procedure A was fol-
lowed by employing PIPYH (50 mg, 0.21 mmol) and Zn(NO3)2 4H2O
3
(0.21 mmol, 2.1 mL of a 0.1 M solution in ethanol) in 10 mL of ethanol;
compound 3was obtained as yellow cubes (54 mg, 78% based on ligand) by
crystallization at room temperature overnight. IR (ATR, cmꢀ1) ν = 1595
(m), 1424 (s), 1585 (s), 1147 (s), 782 (m), 414 (m). EI-MS (m/z) 530
[M ꢀ 2H]+. Anal. Calcd for ZnC20H16N12O6Cl2 0.4(EtOH) 2H2O
3
3
(Found): C, 35.13 (35.13); H, 3.21 (3.20); N, 23.65 (23.62).
EPR spectra were taken on a MS300 (Magnettech, Berlin, Germany)
or Bruker EMX (Bruker, Rheinstetten, Germany) spectrometer. The
samples were kept at 77 K in a dewar finger filled with liquid nitrogen
(the spectra were taken at a modulation amplitude of 5 G). EPR spectra
were simulated with Simfonia (Bruker, Rheinstetten, Germany).
Synthesis of 3-Hydrazino-6-chloropyridazine.33 A solution
of 3,6-dichloropyridazine (2.0 g, 13.4 mmol) in water/NH3 (50 mL) was
heated to reflux when hydrazine hydrate (4.3 g of a 25% solution,
21.5 mmol) was added. Heating was maintained for 3 h, after which full
conversion was evidenced by thin-layer chromatography (TLC). After
the solution was cooled to room temperature, the volume was reduced to
one-third. Upon cooling the solution in an ice bath, the product
precipitated as a beige crystalline material, which was filtered off and
dried in vacuo to afford the product. (1.65 g, 85.2%). 1H NMR (300 MHz,
CD3CN, 300 K) δ = 4.11 (br s, 3H, NH-NH2), 7.12 (d, J = 9.3 Hz, 1H,
General Procedure B (Synthesis of Complexes 4ꢀ7). To a
boiling solution of PIPYH in ethanol was added triethylamine, and the
mixture was refluxed for 5 min. Subsequently 1 equiv of the metal salt
[Co(NO3)2 6H2O, Ni(NO3)2 6H2O, Cu(NO3)2 3H2O, or Zn-
3
3
3
(NO3)2 4H2O] was added as a 0.1 M solution in ethanol. An intensive
3
color change was immediately apparent. The reaction mixtures were
heated to reflux for an additional 20 min, followed by cooling to room
temperature. The products crystallized overnight at ambient tempera-
ture by slow evaporation of the solvent or were precipitated by layering
with diethyl ether. The obtained materials were isolated by filtration or
centrifugation and dried in vacuo.
Synthesis of [CuII(PIPY)(NO3)] (4). General procedure B was followed
by employing PIPYH (50 mg, 0.21 mmol), triethylamine (30 μL,
0.21 mmol), and Cu(NO3)2 3H2O (0.21 mmol, 2.1 mL of a 0.1 M
3
ArHPdz), 7.29 (d, J = 9.3 Hz, 1H, ArHPdz), 7.50 (br s, 1H, NH-NH2). 13
C
solution in ethanol) in 10 mL of ethanol; compound 4 was obtained as
dark brown microcrystalline material (29 mg, 39%) by slow evaporation
of the solvent overnight. The material was washed with chloroform and
dried in vacuo. IR (ATR, cmꢀ1) ν = 1600 (m), 1434 (s), 1397 (s), 1284
(s), 1097 (m), 740 (s), 674 (s), 414 (s). MS (EI, m/z) 295 [M ꢀ NO3]+.
Anal. Calcd for CuC10H7N6O3Cl (Found): C, 31.93 (32.26); H, 2.41
(2.71); N, 22.43 (21.53).
NMR (75 MHz, CDCl3, 300 K) δ = 116.6, 129.2, 145.9, 162.2 (aromatic
carbons).
Synthesis of PIPYH. An ethanolic solution (60 mL) of 3-hydrazi-
no-6-chloropyridazine (1.00 g, 6.9 mmol) and 1.1 equiv of 2-pyridineal-
dehyde (0.73 mL, 7.6 mmol) in the presence of acetic acid (3 drops) as a
catalyst was heated to reflux for 45 min. After the mixture was cooled to
room temperature, the precipitated product was filtered off as white
needles and dried in vacuo to yield 1.21 g (75.1%) of PIPYH. 1H NMR
(300 MHz, CD3CN, 300 K) δ = 11.97 (s, 1H, NH), 8.57 (d, J = 4.9 Hz,
1H, ArHPy), 8.16 (s, 1H, HNdC), 8.03 (d, J = 7.9 Hz, 1H, ArHPy), 7.83
(t, J = 7.7 Hz, 1H, ArHPy), 7.76 (d, J = 9.4 Hz, 1H, ArHPdz), 7.71 (d, J =
9.4 Hz, 1H, ArHPdz), 7.35 (dd, J = 7.1, 4.9 Hz, 1H, ArHPy). 13C NMR (75
MHz, DMSO-d6, 300 K) δ = 116.6, 119.8, 124.0, 130.6, 137.1, 142.8,
148.4, 149.8, 154.0, 159.2. EI-MS (m/z) 233 [M]+. IR (ATR, cmꢀ1) ν =
1576 (m), 1529 (m), 1461 (m), 1407 (s), 1284 (m), 1070 (s), 830 (s),
769 (s), 713 (s), 426 (s).
Synthesis of [NiII(PIPY)(PIPYH)](NO3) (5). General procedure B was
followed by employing PIPYH (50 mg, 0.21 mmol), triethylamine
(30 μL, 0.21 mmol), and Ni(NO3)2 6H2O (0.21 mmol, 2.1 mL of a
3
0.1 M solution in ethanol) in 10 mL of ethanol; compound 5 was
obtained as dark brown microcrystalline material (31 mg, 50% based on
ligand) by slow evaporation of the solvent overnight. The material was
washed with chloroform and dried in vacuo. IR (ATR, cmꢀ1) ν = 1597
(m), 1401 (s), 1290 (s), 1124 (s), 1033 (m), 746 (m), 653 (m). MS (ESI
in CH3CN, m/z) 525 [M]+. Anal. Calcd for NiC20H15N11Cl2O3
CHCl3 (Found): C, 35.71 (35.69); H, 2.28 (2.22); N, 21.81 (22.01).
3
7486
dx.doi.org/10.1021/ic200279g |Inorg. Chem. 2011, 50, 7478–7488

Inorganic Chemistry
ARTICLE
Synthesis of [ZnII(PIPY)2] (6). General procedure B was followed by
employing PIPYH(50mg, 0.21 mmol), triethylamine (30μL, 0.21 mmol),
spectra were corrected for dilution by applying the LambertꢀBeer law
on the supposition that ε remains constant.
and Zn(NO3)2 4H2O (0.21 mmol, 2.1 mL of a 0.1 M solution in
Computational Details. The geometries of [CuII(PIPYH)
(NO3)2] (1) and [CuII(PIPY)(NO3)] (4) complexes used for the
hyperfine structure and g-tensor computations were optimized in
unrestricted KohnꢀSham calculations at the B3LYP44ꢀ46 level by use
of the Turbomole47 package. The basis sets were of polarized triple-ζ
quality for all atoms (TZVP).48 All hyperfine calculations were carried
out with the ORCA49 program package at the optimized geometries and
with the hybrid B3LYP functional. The choice of this functional was
based on previous computations, which show that it is very successful in
the prediction of hyperfine coupling (HFC) and g-tensor in nitrogen
and CuII complexes.50ꢀ53 Ligand atoms were treated by Huzinagaꢀ
Kutzelnigg type basis sets BII (denoted also as IGLO-II).54,55 For Cu
center, an accurate triply polarized basis set CP(PPP) was employed.52
This basis set is especially flexible in the core region and is believed to
provide results close to the basis set limit for the Fermi contact
interaction. Because the spinꢀorbit effects are known to influence the
HFC results for 3d transition metal complexes, in the case of the copper
atom the contributions of spinꢀorbit coupling (SOC) to the HFC were
taken into account. It was proven that implementation of SOC
contributions into the HFC computations results in significant improve-
ment of calculated parameters when compared to experiment.53,56 The
calculation of g-tensor values was carried out with the same geometry
and basis sets. A common gauge at the copper atom was employed.
ORCA package was used also for generating the isosurface plots of spin
density distributions at the B3LYP/TZVP level.
3
ethanol) in 10 mL of ethanol; compound 6 was obtained as an orange
waxy solid (35 mg, 63% based on the ligand) by slow evaporation of the
solvent. 1H NMR (300 MHz, CD3CN, 300 K) δ = 8.29 (s, 1H, HNdC),
7.93 (d, J = 4.6 Hz, 1H, ArHPy), 7.82 (t, J = 7.7 Hz, 1H, ArHPy), 7.46 (d,
J = 7.7 Hz, 1H, ArHPy), 7.17 (t, J = 5.0 Hz, 1H, ArHPy), 7.13 (d, J =
9.5 Hz, 1H, ArHPdz), 7.07 (d, J = 9.4 Hz, 1H, ArHPdz). IR (ATR, cmꢀ1
)
ν = 904 (w), 1018 (w), 1097 (m), 1119 (s), 1294 (m), 1324 (s), 1398 (s),
1593 (m). EI-MS (m/z) 530 [M]+. Anal. Calcd for ZnC20H14N10Cl2
1.5H2O (Found): C, 43.23 (43.11); H, 3.08 (2.99); N, 25.21 (25.41).
Synthesis of [CoIII(PIPY)2](NO3) (7). General procedure B was
followed by employing PIPYH (50 mg, 0.21 mmol), triethylamine
3
(30 μL, 0.21 mmol), and Co(NO3)2 6H2O (0.21 mmol, 2.1 mL of a
3
0.1 M solution in ethanol) in 10 mL of ethanol; compound 7 was
obtained as a dark brown microcrystalline material (44 mg, 71% based
on the ligand) after removal of the solvent. Recrystallization from
ethanol gave green crystals of X-ray quality. 1H NMR (300 MHz,
CD3CN, 300 K) δ = 8.88 (s, 1H, HNdC), 8.04 (t, J = 7.7 Hz, 1H,
ArHPy), 7.93 (d, J = 4.9 Hz, 1H, ArHPy), 7.85 (d, J = 7.7 Hz, 1H, ArHPy),
7.40 (d, J = 9.4 Hz, 1H, ArHPdz), 7.36 (m, 1H, ArHPd), 7.30 (d, J = 9.4
Hz, 1H, ArHPdz). IR (ATR, cmꢀ1) ν = 1539 (m), 1397 (s), 1323 (s),
1119 (s), 1039 (s), 747 (m), 680 (m). MS (ESI in CH3CN, m/z) 523
[M]+. Anal. Calcd for CoC20H14N11Cl2O3 2H2O, brown material
3
(Found): C, 38.60 (39.01); H, 2.92 (2.95); N, 24.76 (24.15). Calcd
for CoC20H14N11Cl2O3 0.5(EtOH) 2H2O, green crystals (Found): C,
3
3
40.21 (39.88); H, 3.05 (2.81); N, 24.56 (24.29).
X-ray Crystallographic Studies. Intensity data sets for com-
pounds 1ꢀ3 and 7 were collected by use of a Bruker Smart APEX II
diffractometer equipped with a graphite monochromator (Mo KR
radiation, λ = 0.710 73 Å) and a charge-coupled device (CCD) detector.
All compounds were measured at 100 K. The structures were solved by
direct methods (SHELXS-97)41 and refined by full-matrix least-squares
techniques against F2 (SHELXL-97).41 The asymmetric units of com-
plexes 1ꢀ3 and 7 contain two metal complex molecules and those of 2
and 3 contain additionally one molecule of ethanol, and 1.5 molecules of
ethanol were found in 7. In the latter, one anion was found to be
disordered over two sites, which were refined with site occupation
factors fixed at 0.5. For the disordered NO3ꢀ anion, the displacement
parameters of the N atom were constrained to be the same and the six
NꢀO distances were restrained to the same value. Two ill-defined
ethanol molecules could not be refined properly; thus the electron
density was removed from the data set with the PLATON SQUEEZE
routine42 and modeled as one fully occupied and one-half occupied
ethanol molecule per metal center. All non-hydrogen atoms were refined
with anisotropic displacement parameters without any constraints. Data
were corrected for absorption and Lp factors with SADABS.43 Hydrogen
atoms were placed geometrically and refined by use of a riding model.
Electrochemical Methods. Redox chemical properties of com-
plexes 1, 2, 4, and 7 were carried out in an argon-filled glovebox by use of
a Gamry Reference 600 potentiostat connected to a personal computer.
Cyclic voltammograms were obtained in 0.05 M solution [nBu4N]
[PF6]/CH3CN at 25 ꢀC. The experiments were carried out in a three-
electrode glass cell with a platinum disk (d = 3 mm) working electrode, a
platinum wire auxiliary electrode, and a [Ag+/AgNO3 0.01 M in
CH3CN, 0.1 M [nBu4N][PF6]] reference electrode. Under these
conditions, oxidation of ferrocene occurs at 0.095 V.
’ ASSOCIATED CONTENT
S
Supporting Information. Three figures showing molec-
b
ular view of complex 3 and Mulliken atomic spin densities for
complexes 1 and 4. This material is available free of charge via the
Internet at http://pubs.acs.org. X-ray crystallographic data for 1,
2, 3 and 7 are also available (CCDC reference numbers 810789,
810788, 810790, and 810791).
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: nadia.moesch@uni-graz.at.
’ ACKNOWLEDGMENT
We acknowledge fForteꢀWissenschafterinnenkolleg FreChe
Materie for financial support. We are grateful to Dr. Christine
Onitsch for help with the EPR spectra.
’ REFERENCES
(1) Zoller, G.; Strobel, H.; Will, D. W.; Wohlfart, P. PCT Int. Appl.
WO 2008 104279 A1, 2008.
(2) Durand-Reville, T.; Jewell, C.; Hammond, C.; Chin, D. PCT Int.
Appl. WO 2008 030579, 2008.
(3) Watterson, D. M.; van Eldik, L.; Haiech, J.; Hibert, M.; Bourguignon,
J.-J.; Velentza, A.; Hu, W.; Zasadzki, M. PCT Int. Appl. WO 2006 050359,
2006.
(4) Trah, S.; Lamberth, C.; Wendeborn, S. V. PCT Int. Appl. WO
2009 090039, 2009.
Protonation Studies. UV/vis spectra were measured on a Varian
Cary 50 spectrophotometer equipped with a thermostated cuvette
holder (25 ꢀC). In a typical experiment, 2 mL of a 0.025 mM solution
of the complex was placed in the photometer. One milliliter of a 2.5 mM
solution of the reagent was added in 10 portions of 100 μL each, which
correspond to 10 equiv. After each addition, a spectrum was acquired. All
(5) Lamberth, C.; Wendeborn, S. V.; Trah, S.; Dumeunier, R. PCT
Int. Appl. WO 2008 089934, 2008.
(6) Grote, T.; Gypser, A.; Rheinheimer, J.; Rose, I.; Schaefer, P.;
Schieweck, F.; Sauter, H.; Gewehr, M.; Mueller, B.; Tormo i Blasco, J.;
Ammermann, E.; Strathmann, S.; Lorenz, G.; Stierl, R. PCT Int. Appl.
WO 2002 074753 A3, 2002.
7487
dx.doi.org/10.1021/ic200279g |Inorg. Chem. 2011, 50, 7478–7488

Inorganic Chemistry
ARTICLE
(7) Deng, Z.; Tseng, H.-W.; Zong, R.; Wang, D.; Thummel, R. P.
Inorg. Chem. 2008, 47, 1835–1848.
(49) Neese, F. ORCA, an ab initio, density functional and semiempirical
program package, University of Bonn, Germany, 2007
(50) Neese, F. J. Phys. Chem. A 2001, 105, 4290–4299.
(51) Kaupp, M.; Reviakine, R.; Malkina, O. L.; Arbuznikov, A.;
Schimmelpfennig, B.; Malkin, V. G. J. Comput. Chem. 2002, 23, 794–803.
(52) Neese, F. Inorg. Chim. Acta 2002, 337, 181–192.
(53) Remenyi, C.; Reviakine, R.; Arbuznikov, A. V.; Vaara, J.; Kaupp,
M. J. Phys. Chem. 2004, 108, 5026–5033.
(54) Kutzelnigg, W.; Schindler, M.; Fleischer, U. In Deuterium and
Shift Calculations; Diehl, P., Fluck, E., G€unther, H., Kosfeld, R., Seelig, J.,
Eds.; NMR Basic Principles and Progress, Vol. 23; Springer: Berlin and
Heidelberg, Germany, 1990
(8) Price, J. R.; Lan, Y.; Brooker, S. Dalton Trans. 2007, 1807–1820.
(9) Ohno, K.; Arima, K.; Tanaka, S.; Yamagata, T.; Tsurugi, H.;
Mashima, K. Organometallics 2009, 28, 3256–3263.
(10) Manzano, B. R.; Jalon, F. A.; Ortiz, I. M.; Soriano, M. L.;
La Gomez Torre, F.; de; Elguero, J.; Maestro, M. A.; Mereiter, K.;
Claridge, T. D. W. Inorg. Chem. 2008, 47, 413–428.
(11) Thompson, L. K.; Woon, T. C.; Murphy, D. B.; Gabe, E. J.; Lee,
F. L.; Le Page, Y. Inorg. Chem. 1985, 24, 4719–4725.
(12) Escuer, A.; Vicente, R.; Mernari, B.; El Gueddi, A.; Pierrot, M.
Inorg. Chem. 1997, 36, 2511–2516.
(13) Brooker, S. Eur. J. Inorg. Chem. 2002, 2535–2547.
(14) Beckmann, U.; Brooker, S. Coord. Chem. Rev. 2003, 245, 17–29.
(15) Allan, J. R.; Barnes, G. A.; Brown, D. H. J. Inorg. Nucl. Chem.
1971, 33, 3765–3771.
(55) Huzinaga, S. Approximate Atomic Functions, University of
Alberta, Canada, 1971
(56) Neese, F. J. Chem. Phys. 2003, 118, 3939–3948.
(16) An, C. X.; Li, X. R.; Zhang, Z. H. Transition Met. Chem. 2009,
34, 255–261.
(17) Seleem, H. S. J. Indian Chem. Soc. 2003, 80, 622–625.
(18) Blake, A. J.; Hubberstey, P.; Li, W.-S.; E.Russell, C.; Smith, B. J.;
Wraith, L. D. Inorg. Chem. 1998, 4, 647–656.
(19) Mahmoud, F. Z.; Ramadan, A. A. T.; Ali, D. A. J. Coord. Chem.
2008, 61, 2639–2654.
(20) Gr€unwald, K. R.; Saischek, G.; Volpe, M.; Belaj, F.;
M€osch-Zanetti, N. C. Eur. J. Inorg. Chem. 2010, 15, 2297–2305.
(21) Lions, F.; Martin, K. V. J. Am. Chem. Soc. 1957, 80, 3858–3865.
(22) Green, R. W.; Hallman, P. S.; Lions, F. Inorg. Chem. 1963,
3, 376–381.
(23) Chiswell, B.; Geldard, J. F.; Phillip, A. T.; Lions, F. Inorg. Chem.
1964, 3, 1272–1277.
(24) Mesa, J. L.; Arriortua, M. I.; Lezama, L.; Pizarro, J. L.; Rojo, T.;
Beltran, D. Polyhedron 1988, 7, 1383–1388.
(25) Rojo, T.; Mesa, J. L.; Arriortua, M. I.; Savariault, J. M.; Galy, J.;
Villeneuve, G.; Beltran, D. Inorg. Chem. 1988, 27, 3904–3911.
(26) Gerloch, M. J. Chem. Soc. A 1966, 1317–1325.
(27) Geldard, J. F.; Lions, F. Inorg. Chem. 1963, 2, 270–282.
(28) Bell, C. F.; Quddus, M. A. Anal. Chim. Acta 1970, 52, 313–321.
(29) Cameron, A. J.; Gibson, N. A. Anal. Chim. Acta 1970, 51,
249–256.
(30) Quddus, M. A.; Bell, C. F. J. Inorg. Nucl. Chem. 1971, 33,
2001–2007.
(31) Quddus, M. A.; Bell, C. F. Anal. Chim. Acta 1968, 42, 503–513.
(32) Heit, M. L.; Ryan, D. E. Anal. Chim. Acta 1965, 32, 448–455.
(33) Farina, C.; Pinza, M.; Cerri, A.; Parravinci, F. PCT Int. Appl. EP
274409 A2, 1988.
(34) Kime, N. E.; Ibers, J. A. Acta Crystallogr. 1969, B25, 168–169.
(35) Otieno, T.; Rettig, S. J.; Thompson, R. C.; Trotter, J. Inorg.
Chem. 1995, 34, 1718–1725.
(36) Dumitru, F.; Petit, E.; van der Lee, A.; Barboiu, M. Eur. J. Inorg.
Chem. 2005, 4255–4262.
(37) Dumitru, F.; Legrand, Y.-M.; Barboiu, M.; Petit, E.; van der Lee,
A. Cryst. Growth Des. 2009, 9, 2917–2921.
(38) Sarkar, B.; Konar, S.; Gomez-Garcia, C. J.; Ghosh, A. Inorg.
Chem. 2008, 47, 11611–11619.
(39) Prasad, K. T.; Scaife, D. B. J. Electroanal. Chem. 1977, 84, 373–386.
(40) Brodovitch, J. C.; Haines, R. I.; McAuley, A. Can. J. Chem. 1981,
59, 1610–1614.
(41) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112–122.
(42) van der Sluis, P.; Spek, A. L. Acta Crystallogr. 1990, A 46,
194–201.
(43) SADABS, v2008-1; Bruker AXS, 2008.
(44) Becke, A. D. Phys. Rev. A 1988, 38, 3098–3100.
(45) Becke, A. D. J. Chem. Phys. 1993, 98, 1372–1377.
(46) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652.
(47) Ahlrichs, R.; Bar, M.; Haser, M.; Horn, H.; K€olmel, C. Chem.
Phys. Lett. 1989, 162, 165–169.
(48) Sch€afer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
5829–5835.
7488
dx.doi.org/10.1021/ic200279g |Inorg. Chem. 2011, 50, 7478–7488